paper-with-me

Papers

VLM Judges Can Rank but Cannot Score: Task-Dependent Uncertainty in Multimodal Evaluation

2026-04-28 · Divake Kumar, Sina Tayebati, Devashri Naik, Ranganath Krishnan, Amit Ranjan Trivedi arxiv

Vision-language models (VLMs) are increasingly used as automated judges for multimodal systems, yet their scores provide no indication of reliability. We study this problem through conformal prediction, a distribution-free framework that converts a judge's point score into a calibrated prediction interval using only score-token log-probabilities, with no retraining. We present the first systematic analysis of conformal prediction for VLM-as-a-Judge across 3 judges and 14 visual task categories. Our results show that evaluation uncertainty is strongly task-dependent: intervals cover ~40% of the score range for aesthetics and natural images but expand to ~70% for chart and mathematical reasoning, yielding a quantitative reliability map for multimodal evaluation. We further identify a failure mode not captured by standard evaluation metrics, ranking-scoring decoupling, where judges achieve high ranking correlation while producing wide, uninformative intervals, correctly ordering responses but failing to assign reliable absolute scores. Finally, we show that interval width is driven primarily by task difficulty and annotation quality, i.e., the same judge and method yield 4.5x narrower intervals on a clean, multi-annotator captioning benchmark. Code: https://github.com/divake/VLM-Judge-Uncertainty

📄 PDF Abstract BibTeX arXiv:2604.25235

Code (0)

등록된 구현이 없습니다.

Tasks

Mathematical Reasoning

Similar Papers 제목 키워드 기반

JuStRank: Benchmarking LLM Judges for System Ranking

2024-12-12 · Ariel Gera, Odellia Boni, Yotam Perlitz, Roy Bar-Haim 외

Given the rapid progress of generative AI, there is a pressing need to systematically compare and choose between the numerous models and configurations available. The scale and versatility of such evaluations make the us…

Benchmarking

CyclicJudge: Mitigating Judge Bias Efficiently in LLM-based Evaluation

2026-03-02 · Ziyi Zhu, Olivier Tieleman, Alexey Bukhtiyarov, Jinghong Chen arxiv

LLM-as-judge evaluation has become standard practice for open-ended model assessment; however, judges exhibit systematic biases that cannot be averaged out by increasing the number of scenarios or generations. These bias…

Evaluative Fingerprints: Stable and Systematic Differences in LLM Evaluator Behavior

2026-01-08 · Wajid Nasser arxiv

LLM-as-judge systems promise scalable, consistent evaluation. We find the opposite: judges are consistent, but not with each other; they are consistent with themselves. Across 3,240 evaluations (9 judges x 120 unique vid…

R$^3$-SQL: Ranking Reward and Resampling for Text-to-SQL

2026-04-28 · Hojae Han, Yeonseok Jeong, Seung-won Hwang, Zhewei Yao 외 arxiv

Modern Text-to-SQL systems generate multiple candidate SQL queries and rank them to judge a final prediction. However, existing methods face two limitations. First, they often score functionally equivalent SQL queries in…

Re-Rankers as Relevance Judges

2026-01-08 · Chuan Meng, Jiqun Liu, Mohammad Aliannejadi, Fengran Mo 외 arxiv

Using large language models (LLMs) to predict relevance judgments has shown promising results. Most studies treat this task as a distinct research line, e.g., focusing on prompt design for predicting relevance labels giv…